Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Oil immersion cooling can reduce data-center cooling energy, but it does not guarantee a fixed percentage of savings. In single-phase immersion, servers sit in a nonconductive liquid that stays liquid as it absorbs heat; pumps and heat exchangers then move that heat to facility cooling equipment. Two-phase systems use a fluid that boils at hot components and condenses back into liquid. Whether either design saves energy depends on the facility, climate, heat-rejection system, IT load, and what the comparison counts.
What oil immersion cooling means
In a single-phase immersion system, server equipment is submerged in a dielectric liquid. The liquid absorbs heat without boiling, then circulates through a heat exchanger so the heat can be rejected elsewhere. A 2021 system-level experiment described its single-phase system as using oil; that does not mean every immersion system uses ordinary oil or that one fluid suits every server and tank. The experiment compared direct-immersion data-center units.
Two-phase immersion uses a working fluid that boils near heated equipment. The vapor rises and condenses back into liquid in the system. In the same 2021 experiment, the two-phase system used an engineered dielectric liquid, not the oil used in the single-phase system. “Oil cooling” is therefore an imprecise shorthand for a broader set of immersion technologies.
These systems differ from direct-to-chip liquid cooling, where liquid passes through cold plates attached to components such as CPUs or GPUs. Other server parts may remain air-cooled. Traditional air cooling uses server fans and room or facility equipment to remove heat.
#1 Best Overall
- Transceivers/AOCs Solution for Liquid Immersion Cooling System;
- Use 3M flexible cable with nylon jacket;
- Insertion Loss 7.5dB;
- Oil-resistance, dust-proof and anti-corrosion;
- Customizable EEPROM mapping.
Does immersion cooling save energy?
It can, but the result depends on the system design and site. A 2026 study comparing high-density data centers reported annual PUE 0.078 lower for immersion cooling with a water-side economizer than for air cooling in the configurations it studied. The study analyzed different climates and economizer arrangements; its result is not a universal prediction for every facility. Read the study’s air-versus-immersion comparison.
PUE, or power usage effectiveness, is a ratio of total data-center energy to IT-equipment energy. A lower PUE indicates less overhead relative to the computing load, but it does not by itself say how much water a facility uses, what its emissions are, or whether it reuses waste heat.
A 2021 controlled system-level experiment found nearly 75% better coefficient-of-performance and a 5.1% better PUE trend for its tested two-phase system than for its tested single-phase system. Those are comparisons between the specific systems and operating ranges tested, not a general ranking of all two-phase and single-phase products. See the experimental study.
Free tools Windows power users keep installed
One-click scans. No signup required.
What other liquid-cooling savings figures do—and do not—show
Liquid cooling more broadly has evidence of potential efficiency gains, but direct-to-chip results should not be presented as oil-immersion results. A 2026 study developed an energy model validated against on-site measurements at a Melbourne data center, then used it to predict outcomes under other conditions. For air-to-chip to liquid-to-chip conversion, it modeled 4%–13% lower annual energy use, emissions, and PUE per unit of compute, and 6%–14% lower peak power demand per unit of compute. These are modeled outcomes under the paper’s assumptions, not measured immersion savings. Read the liquid-to-chip study.
That paper also modeled a control approach that increased the liquid-to-chip differential temperature from 5 °C to 10 °C. It reported PUE falling from a modeled 1.22–1.25 range to 1.18, associated with about 3%–6% total-facility efficiency improvement and 18%–28% potential central-plant energy reduction. These figures describe the study’s proposed control method and modeled conditions, not results from an oil-immersion installation.
For a different boundary, a 2020 Electric Power Research Institute laboratory evaluation measured a 14% overall data-center energy reduction for one negative-pressure direct-to-chip setup. EPRI said production-scale testing was needed. This is neither an immersion result nor a universal expectation for liquid cooling. Read EPRI’s evaluation.
How to compare cooling choices fairly
A useful comparison names both the cooling architecture and the boundary of the energy figure. Server-only power, cooling-system power, central-plant energy, and total facility energy are different measures; a percentage for one cannot be compared directly with a percentage for another.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problems- Air cooling: Heat is carried by server fans and room or facility cooling equipment. The relevant comparison includes both IT and cooling loads.
- Single-phase oil immersion: Equipment is submerged in a liquid that remains liquid. Pumps, heat exchangers, and the facility’s heat-rejection system still use energy.
- Two-phase immersion: A dielectric fluid boils at hot components and condenses in the system. The 2021 test favored its tested two-phase unit on COP and PUE trends, but did not establish a universal product advantage.
- Direct-to-chip cold plates: Liquid cools selected components rather than submerging the entire server. The 2026 liquid-to-chip model and EPRI’s 2020 direct-to-chip test concern this architecture, not immersion.
Climate, IT load, rack density, economizer choice, operating temperatures, and auxiliary equipment all affect the outcome. The 2026 immersion comparison found that economizer arrangements mattered; the liquid-to-chip study modeled effects from changing temperature differential. A facility should compare like-for-like workloads and weather conditions and state whether the result is modeled, measured, or projected.
Energy is not the whole sustainability picture
A 2025 life-cycle analysis of advanced data-center cooling scenarios, including cold plates and immersion, reported 15%–20% lower energy demand, 15%–21% lower greenhouse-gas emissions, and 31%–52% lower blue-water consumption across evaluated alternatives. These are life-cycle scenario results, not guaranteed reductions at a particular site. Read the life-cycle analysis.
For a site-level decision, examine electricity, water, emissions, and any useful heat recovery using a consistent boundary. Do not assume immersion is waterless or categorically better for sustainability: heat still has to leave the system, and the result depends on the heat-rejection design and the life-cycle factors being counted.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What operators should check before adopting immersion
Efficiency is only one part of an operational decision. Immersion changes how hardware is installed, serviced, and supported, and those practical requirements can affect both risk and cost.
- Compatibility: Confirm that servers, components, materials, and fluid are suitable together. Do not assume a standard server can be submerged without manufacturer and system-provider approval.
- Service and fluid handling: Establish procedures for removing, servicing, and returning equipment, and for managing the fluid. Operators need to know how maintenance differs from their existing air-cooled workflow.
- Leak and reliability controls: Assess containment, monitoring, and response procedures as part of system design. EPRI’s 2020 assessment identified compatibility and perceived or actual leak risk as adoption barriers in the then-available market.
- Support and warranties: Verify current hardware-maker support and warranty terms for the exact configuration. EPRI’s 2020 report is a technical baseline, not a current vendor catalog.
- Whole-system economics: Include equipment, facility modifications, operating energy, maintenance, and the chosen heat-rejection system. The available studies do not establish a reliable universal cost ranking.
So, deep-fried or deep energy savings?
Immersion cooling is a real engineering option that can lower cooling overhead and support lower facility energy use in suitable systems. The 2026 air-versus-immersion comparison supports that case for its studied configurations, while experimental and modeling results show why the size of any gain depends on architecture and operating conditions. Treat headline percentages as evidence from a specific study—not as a savings promise for every data center.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
